The invention relates to the field of electrical power converters, and more specifically, to AC-DC conversion circuits using a principle of intermittently charging a capacitor.
Such converters are notably used directly for motorized products in buildings, such as small screens for solar protection, venetian blinds, etc. They make it possible to convert the AC voltage from the power mains (xe2x80x9cmainsxe2x80x9d) into a lower virtually DC voltage capable of being used to supply an actuator with a low-power DC motor, for example of the CONCEPT-25 type product manufactured by the applicant. They may also supply only the control electronics of a more powerful AC motor, for example, for roller shutters, or a single sensor combined with a radio frequency receiver or transmitter.
These converters are termed xe2x80x9cunregulatedxe2x80x9d because they operate without their output voltage being tied to a fixed reference. This simplifies the structure of the converter and lowers the cost thereof At the input, these converters receive an AC current, which may or may not be rectified; the half-wave or full-wave rectification of an AC current is considered in the present description as known in the art and is not described. At the output, these converters provide a virtual DC voltage with an amplitude less than the peak amplitude of the input voltage. In the remainder of the description, it is assumed that the input voltage is the mains voltage. This assumption is the most common for applications in buildings. It is clear, however, that other applications are possible, and that the input AC current of the converter, whether or not rectified, is not limited to the current supplied by the mains. These converters are called low-power converters, in that it is possible to tolerate therein a lower conversion efficiency than in a high-power plant, and therefore proportionally larger losses. The output power is typically less than 50 W.
In principle, as explained for example in FIG. 1 of U.S. Pat. No. 4,641,233, the converters use the intermittent charging of a high-capacitance capacitor, hereinafter called C. In this context, the term xe2x80x9chigh capacitancexe2x80x9d refers to capacitances which are often greater than 100 xcexcF. However, it will be understood that the capacitance of the capacitor used is simply sufficient for maintaining enough output voltage between the periods of intermittently charging the capacitor. As a result, the capacitance of the capacitor depends on the variation in voltage accepted at the output of the voltage converter, on the output current needed, and on the charging frequency, which itself depends on the input voltage frequency. For low output currents or for a high frequency of the input voltage, a capacitor with a lower capacitance at this threshold of 100 xcexcF could be used. A switching element, hereinafter called Q, is placed between the half-wave or full-wave rectified mains and this capacitor C, so as to charge the latter only during time periods when the mains voltage remains below a given threshold. For a given load, it is thus possible to dimension C and to choose the threshold voltage so as to comply with a given service voltage and a given waveform at the output of the converter.
U.S. Pat. No. 4,001,668 (Lewis 1973) describes a device of this sort in FIG. 4. This patent emphasizes the very high tolerance to variations in the mains voltage, allowing the use of such a circuit to supply a motor with DC current whether power is provided by 110-120 V AC mains or from 220-240 V AC mains.
The conducting and non-conducting states of the switching element Q (47) are completely determined by comparison of a voltage, which is an image of the rectified voltage, with a threshold (50, 51, 45). It should be noted that the use of a current limiter formed by a resistor RP (48) and placed upstream of the voltage measuring point makes it possible to benefit from a cumulative effect favoring fast switching of Q from an off state to a conducting state or vice-versa. In FIGS. 5 and 6, this patent clearly indicates the pulsed shape of the line current (61, 65) and of the downstream voltage waveform at the terminals of the load (therefore at the terminals of C) (62, 66). Patents filed subsequently constitute particular embodiments making it easier to control the switch Q, or improvements relating to the reduction of harmonics generated by current spikes necessary for the periodic recharging of the capacitor.
GB-A-2 203 003 (Sanderson 1987) describes, in generic terms, a topology which is slightly different but based on the same principle. In FIG. 2, a current limiting device (11) is inserted between a switching element formed by a field effect transistor and the capacitor (6), the voltage measurement for controlling the switching element being carried out on C (by means of the block referenced 10). This arrangement may remove the possibility of a cumulative effect for the effective control of Q. A device (12) is intended to take the switching element out of service when the voltage at its terminals exceeds a maximum threshold, however, this device is not described.
EP-A-0 763 878 (Helfrich 1995) describes, in FIG. 1, a converter topology using a current limiting device RP (R1), a switching element Q (Q1) and a capacitor C (C2) with composite control of Q both by the voltage upstream of RP (R6) and by the downstream voltage (R6). However, the voltage threshold is the same for the conducting and non-conducting state of the switching element.
Other embodiments of converters using this intermittent charge principle are described in DE-A-44 44612, DE-A-31 44742, EP-A-0 399 598, DE-A-32 45238, EP-A-0 249 259, FR-A-2 672 448, EP-A-0 500 113.
EP-A-0 622 889 (Wong 1994) describes a series topology (Q, C) without a current limiter. The input voltage Vinxe2x80x94the voltage of the rectified mainsxe2x80x94is applied to the series circuit of the capacitor and of the switching element. This circuit is intended to allow, as in the previous cases, double charging of C per half-cycle, and therefore to reduce the waveform upstream of a regulator. The device analyzes both the voltage at the terminals of C (15) and the voltage at the terminals of Q (16). Q becomes conducting if the voltage Vcap at the terminals of the capacitor is less than a first given threshold. Q also becomes conducting if the voltage Vds at the terminals of Q, which is the difference between the rectified input voltage and the voltage at the terminals of the capacitor, becomes less than a second given threshold. In one half-cycle, this allows the capacitor to be charged when the voltage increases from zero, then the capacitor to be charged when the voltage decreases to zero.
U.S. Pat. No. 4,641,233 (Roy 1985) describes, with reference to FIG. 1, a similar topology. The switching device, a bipolar transistor, becomes conducting if the output voltage of the converter (downstream voltage) is less than the chosen reference voltage. However, the switching device only becomes conducting if the voltage upstream of the switching element is less than a voltage, which is about twenty volts for an output voltage of five volts. This is obtained using a second transistor, which is switched on according to the input voltage, and which switches off the first transistor when the input voltage is too high. According to the patent, the objective of the second transistor is to prevent damaging the components as a result of switching at a high voltage.
A drawback of this device is that the threshold values for the input voltage and the output voltage are only approximately set. It is known that the conducting threshold voltage of a bipolar transistor or of a single p-n junction are not accurately defined. The proposed solution involves replacing the second transistor by a precision reference-voltage source.
There is therefore a need for a converter, which makes it possible for the switching element to be accurately controlled, allowing the capacitor to be charged by the input voltage.
A second drawback of the circuit proposed by Roy is that it artificially limits the dynamic operating range at the output. This is because this circuit prevents the switching element from conducting as soon as the upstream voltage exceeds a fixed threshold, although it is quite possible that when this threshold is reached, the capacitor C is not yet sufficiently recharged. This effect will occur on the rising edges of the rectified sinusoid. It is therefore beneficial in order to prevent limiting the dynamic operating range of the circuit to choose a high threshold for the upstream voltage.
A third drawback of the circuit proposed by Roy appears on the falling edges of the rectified sinusoid. This is because if the capacitor is discharged, the switching element becomes conducting as soon as the upstream voltage goes below the fixed threshold. If this threshold for the upstream voltage is high it becomes conducting at a higher voltage when going back below the threshold, this time during the falling edge of the rectified sinusoid. This results in an operation which is more damaging to the component, and therefore in overdimensioning thereof, and a risk of incompatibility with the electromagnetic compatibility standards. It is therefore beneficial to choose a low threshold for the upstream voltage.
The invention is based on the demonstration of these contradictory requirements in Roy""s circuit and on the discovery of the corresponding problem. There is therefore a need for a converter which does not limit the dynamic operating range at the output, but which nevertheless allows the components to be protected, and which limits the risks of electromagnetic incompatibility. It would also be beneficial for the converter to be able to measure the consumption of a device which it supplies.
Moreover, it may be necessary to transmit a single signal over an electrical line for supplying actuator(s) or sensor(s), which in particular, may involve an alarm signal, a fault signal, or an acquisition signal in the actuator configuration phase, or a security signal for a sensor or the like. It is of course possible to use a medium other than the electrical supply line, for example radio transmission, or a specific conductor for transmitting the signal. However, it is beneficial to use the electrical line, which avoids providing another medium. It has been proposed in U.S. Pat. No. 3,852,740 (Haynes 1973), U.S. Pat. No. 4,121,201 (Weathers 1974) or else in U.S. Pat. No. 4,755,792 (Pezzolo 1987) to use carrier current technologies to transmit a signal over a supply line. These technologies are complex and their reliability is not guaranteed if carried out at low cost. They are not suitable for very limited data transmission. There is therefore a need for simple and reliable transmission of a signal, over the supply line of an actuator or a sensor.
FR-A-2 785 735 discloses a voltage converter. The converter has a capacitor, which is charged at most once per half-wave of the rectified input AC voltage, close to the beginning of the half-wave. Specifically, the rectified input AC voltage is applied to the capacitor through a switch. The switch is controlled to be opened whenever the output voltage across the capacitor is higher than a first threshold. The switch is controlled to be closed whenever the difference between the input voltage and the output voltage is higher than a second threshold, which may be zero. Thus, the switch is closed not exclusively based on the input voltage, but rather according to the difference between the input voltage and the output voltage. This may cause damage to the components of the converter if the switch is closed while the input voltage is high.
U.S. Pat. No. 5,818,708 discloses a voltage converter. Rectified input AC voltage is applied to a load capacitor through a switch. The switch is controlled by a latch circuit. The switch is turned on upon receiving a set input, and is turned off upon receiving a reset input. The set input of the latch is connected to a first voltage sensor, which senses the voltage provided by the rectifier. The reset input of the latch is connected to a second voltage sensor, which senses the voltage across the load capacitor. The first voltage sensor enables the set input of the latch when the voltage provided by the rectifier is higher than a low (typically zero) voltage. This turns the switch on. The second voltage sensor enables the reset input of the latch when the voltage provided by the rectifier is higher than the desired output voltage, which turns the switch off. In this converter, the capacitor is exclusively charged on the rising edge of the input voltage.
In one embodiment, the invention therefore provides a voltage converter with input terminals, a series circuit between the input terminals, the series circuit comprising a switching element and a capacitor. The discharge of the capacitor supplies the output voltage of the converter. The circuit also comprises a control circuit of the switching element, which controls the conducting state of the switching element exclusively as a function of the input voltage and controls the non-conducting state of the switching element exclusively as a function of the output voltage.
Advantageously, the control circuit controls the conducting state of the switching element when the input voltage is less than a first threshold, and controls the non-conducting state of the switching element when the output voltage is greater than a second threshold. In this case, the first threshold may be higher than the second threshold so that the capacitor is charged on both rising and falling edges of the input voltage.
In another embodiment, the invention therefore provides a voltage converter with input terminals, a series circuit between the input terminals, the series circuit comprising a switching element and a capacitor. The discharge of the capacitor supplies the output voltage of the converter. The circuit also comprises a control circuit of the switching element. This circuit sets the switching element to the conducting state when the input voltage is less than a first threshold. It also sets the switching element to the non-conducting state when the output voltage is greater than a second threshold, which is lower than first threshold